ӳ����ý

By analyzing tens of thousands of genes, proteins, and protein modifications in hundreds of lung tumors, an international research team has uncovered new factors linked to poor outcomes in lung adenocarcinoma in both smokers and people who have never smoked. These factors include a pattern of DNA damage and signatures of exposures to chemicals found in processed foods and cosmetics. The findings could lead to better precision medicine strategies for this most common type of lung cancer. 

The researchers systematically studied genes and proteins across the whole genome and proteome in lung tumors and paired noncancerous lung tissues from more than 400 patients from North America, Eastern Europe, and Asia, making it the largest “proteogenomic” study of any cancer to date. 

By creating and studying this detailed molecular portrait of lung adenocarcinoma, the team found: 

• several candidate biomarkers that could be used to more accurately identify patients with poor prognoses
• a strong link between poor outcomes and a type of DNA damage in tumors in which the DNA was broken up into many small fragments
• biological mechanisms underlying a subset of early-stage tumors that behave more like advanced cancers
• more than two dozen tumor molecules that could potentially be targeted by existing FDA-approved drugs, compounds in clinical trials, or new medicines yet to be developed. 

The work, published in , was led by scientists from the ӳ����ý, Academia Sinica in Taiwan, the University of Michigan in Ann Arbor, and other institutions, and is the result of an unprecedented collaboration between the and the , which brings together researchers from more than a dozen countries. 

Because of this collaboration, the team was able to access a large number of tumor samples and deeply analyze a range of lung cancer risk factors that haven’t been well studied before, such as exposures to chemicals beyond tobacco smoke. 

“Lung cancer continues to be the leading cancer killer worldwide, and even though we've made progress, it is not a solved problem,” said Michael Gillette, co-senior author of the study and senior group leader and principal investigator in the Proteomics Platform at the ӳ����ý. “My hope is that by bringing new understanding to the biology of lung adenocarcinoma, proposing some therapeutic strategies and new ways to identify high-risk patients, and also by providing an important resource for researchers, our work will lay the groundwork for better ways to diagnose and treat this deadly cancer.”

“Large-scale collaborations such as this one enable access to genetically and geographically varied sample sets, making this study more truly reflective of the diversity of lung adenocarcinoma," said Shankha Satpathy, co-first author of the study and a senior group leader and principal investigator in the ӳ����ý’s Proteomics Platform when the study was conducted.

Lung cancer atlas

Despite new targeted therapies for lung cancer over the last two decades, lung adenocarcinoma still kills the majority of patients, including many who have never smoked, suggesting the need for a deeper look at the disease. 

In their study, the researchers found that patients who had tumors with DNA that was shattered into many small pieces had worse prognoses. The scientists pinpointed a potential biomarker, a protein called IGF2BP3, whose levels correlated with the intensity of this genome fragmentation. The team said this candidate biomarker could potentially be used to identify patients with increased risk of severe disease.

Because of the study’s large sample size, the scientists were able to delve into the effects of carcinogens such as tobacco smoke, air pollutants, and chemicals called nitrosamines found in processed food and cosmetics. The researchers looked for molecular signatures associated with smoking and nitrosamine exposures, and found that patients who had the strongest nitrosamine signatures were mostly never-smokers who nevertheless fared as poorly as those who had smoked. 

This surprising finding suggests that nitrosamines may be an important contributor to lung cancer risk in people who have never smoked. The team created a panel of biomarker candidates that could potentially provide prognostic signals for the disease in people with high chemical exposures. 

The scientists also learned more about a recently discovered subgroup of lung adenocarcinoma, where patients have tumors that look like early-stage disease according to conventional measures but have molecular characteristics and outcomes of late-stage cancer. The researchers revealed new details about the underlying biology fueling this “early-stage, late-like” form and suggested that these patients might benefit from closer monitoring and more intensive treatment.

Gillette said the study shows the power of international collaboration and large-scale genomic and proteomic analysis of cancer, which may pave the way toward better clinical care. 

“The way we do prognostication now is based on clinical measures and how tumor biopsies look under a microscope,” said Gillette. “But if we can move towards using molecular signatures from tumors to determine cancer prognosis, we could do a better job of finding those early-stage cancers that will turn aggressive and treating those patients earlier and more effectively to save more lives.”    

  
The other first authors of the study are Natalie Clark of the ӳ����ý, Yi-Ju Chen of Academia Sinica in Taiwan, Noshad Hosseini and Yi Hsiao of University of Michigan in Ann Arbor, Ya-Hsuan Chang of National Health Research Institutes in Taiwan, Jonathan Lei of Baylor College of Medicine in Houston, Texas, Francesca Petralia of Icahn School of Medicine at Mount Sinai in New York, and Jin-Shing Chen of National Taiwan University Hospital and National Taiwan University College of Medicine. 

The study’s other senior authors are Yu-Ju Chen of Academia Sinica, Marcin Cieslik of University of Michigan, Ann Arbor, and D. R. Mani of the ӳ����ý.